Wenhao Chai

Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces

MM-HELIX: Boosting Multimodal Long-Chain Reflective Reasoning with Holistic Platform and Adaptive Hybrid Policy Optimization

VideoNSA: Native Sparse Attention Scales Video Understanding

Beyond Text-to-Image: Liberating Generation with a Unified Discrete Diffusion Model

AutoCode: LLMs as Problem Setters for Competitive Programming

UEval: A Real-World Benchmark for Unified Multimodal Generation

Learning Text-driven 3D Human Motion Generation from 3D-free Web Videos

Dense Video Understanding with Gated Residual Tokenization

GAM-Agent: Game-Theoretic and Uncertainty-Aware Collaboration for Complex Visual Reasoning

AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark

PAD: Personalized Alignment of LLMs at Decoding-time

HomieBot: an Adaptive System for Embodied Mobile Manipulation in Open Environments